'experimental uncertainty'?

[August]

Hi all

 

I wonder what 'experimental uncertainty' means here.

 

'All measurements are approximations--no measuring device can give perfect measurements without experimental uncertainty.'

 

Thank you

[recursive]

Experimental uncertainty just means that you can't know every aspect of your experiment. If you did, Heisenberg would rise from his grave and get very angry with you...

[Cap'n Refsmmat]

Your thermometer gives you a reading of 35.3 degrees. Is it 35.31? 35.34? You don't know. The measuring device is only approximating the temperature.

 

http://en.wikipedia.org/wiki/Heisenberg_uncertainty_principle

[D H]

Experimental uncertainty just means that you can't know every aspect of your experiment. If you did, Heisenberg would rise from his grave and get very angry with you...

That is the lay interpretation of the uncertainty principle, and it is essentially incorrect. The uncertainty principle talks about measurements of a pair of canonical variables. Variables whose product has units of energy are canonical variables For example, position and momentum. You can know one to any degree of accuracy you want at the expense of the other.

 

August, the answer does not lie in the uncertainty principle. It lies in noise, interference, bias, technique, perception, etc. Here's an experiment for you to conduct on experimental error. Put two black spots about six inches apart on a white sheet of paper. (Don't make it exactly six inches; this will introduce bias.) Get a ruler that has demarcations in tenths of inches on one edge and 1/8 inches on another. Your classmates are to measure the distance between the dots using a ruler that you supply and write their measurements down on slips of paper. Do not tell them details, such as how to measure the distance or which edge of the ruler to use. Some people will be sloppy, some precise. Some will measure from center to center, others from outer edge to outer edge, and so on. Some will use the 1/8" scale, others the 1/10" scale. All of these factors, plus the inherent limitations in using a ruler to estimate distance contribute to experimental error.

[thedarkshade]

Experimental uncertainty (probably) means being unable to get all the certain informations for everything that happens in that particular experiment. It might be the instrumentation, differences in experimentation procedure done by individuals and other factors that eventually effect and make it be called "uncertain".

[August]

Thank you for all your replies. DH, the experiment you suggested was really helpful and interesting. Thanks ; )

 

August

[Theophrastus]

In regards to experimental uncertainty, I have always thought of it in the context of measurement, as we are all human, it is only human nature for our measurements to be correct, to a relative degree. Scientists often like to take note of the theoretical uncertainty of a certain measurement. For example, (doing so with a completely open mind) use a 30cm ruler, to measure the circumference of a door, to the nearest mm. Repeating this experiment several times, one will find that the various measurements shall not be absolutely coherent with one another, but shall be, perhaps, between one and two cm's long or short of one another. Thus experimental uncertainty. In regards to the constant mention of the work of Heisenberg, I disagree in throwing him into this arguement, as he proved that at a quantum mechanical level, it is impossible to make accurate measurements of speed, length, distance, etc., he was not refering to the macroscopic world, however, I admit that they may bear implications regarding measurement, that transcend his actual context.